Neural competition for motion segmentation
نویسندگان
چکیده
We present a system for sensory classification and segmentation of motion trajectories. It consists of a combination of manifolds from Unsupervised Kernel Regression (UKR) and the recurrent neural Competitive Layer Model (CLM). The UKR manifolds hold learned representations of a set of candidate motions and the CLM dynamics, working on features defined in the UKR domain, realises the segmentation of observed trajectory data according to the competing candidates. The evaluation on trajectories describing four different letters yields improved classification results compared to our previous, pure manifold approach.
منابع مشابه
Layered Motion Segmentation with a Competitive Recurrent Network
Using local motion information data such as that obtained from optical flow, we present a network for a multilayered segmentation into motion regions that are governed by affine motion patterns. Using an energy-based competitive multilayer architecture based on non-negative activations and multiplicative update rules, we show how the network can perform segmentation tasks that require a combina...
متن کاملNeural Competitive Structures for Segmentation Based on Motion Features
Simoncelli & Heeger studied how the motion is processed in humans (V1 and MT areas) and proposed a model based on neural populations that extract the local motion structure through local competition of MT like cells. In this paper we present a neural structure that works as dynamic filter on the top of this MT layer and can take advantage of the neural population coding that is supposed to be p...
متن کاملNeural models of motion integration and segmentation
A neural model is developed of how motion integration and segmentation processes compute global motion percepts. Figure-ground properties, such as occlusion, influence which motion signals determine the percept. For visible apertures, a line's extrinsic terminators do not specify true line motion. For invisible apertures, a line's intrinsic terminators create veridical feature tracking signals,...
متن کاملBrain Inspired Cognitive Systems
Bio-inspired energy models compute motion following the guidelines suggested by the neurophysiological studies of V1 and MT areas of monkeys and human behaviour that use neural populations to extract the local motion structure through local competition of MT like cells. In this paper we present a neural structure that works as dynamic filter on the top of this MT layer for image segmentation an...
متن کاملBio-inspired Motion-Based Object Segmentation
Although motion extraction requires high computational resources and normally produces very noisy patterns in real sequences, it provides useful cues to achieve an efficient segmentation of independent moving objects. Our goal is to employ basic knowledge about biological vision systems to address this problem. We use the Reichardt motion detectors as first extraction primitive to characterize ...
متن کامل